Cost-Effectiveness Of A Peer Intervention For The Treatment Of Latent TB Infection In Harlem

Paul W. Colson
Charles P. Felton National TB Center
2238 Fifth Ave., 1st Floor
New York, NY  10037
(212) 939-8241
FAX: (212) 939-8259
pwc2@columbia.edu
 
S Glied
Division of Health Policy & Management
Mailman School of Public Health
Columbia University
601 W. 168th St.
New York, NY  10032
 
W El-Sadr
Charles P. Felton National TB Center
2238 Fifth Ave., 1st Floor
New York, NY  10037
 
Y Hirsch-Moverman
Charles P. Felton National TB Center
2238 Fifth Ave., 1st Floor
New York, NY  10037
 
D Ferry
Division of Health Policy & Management
Mailman School of Public Health
Columbia University
601 W. 168th St.
New York, NY  10032
 
G Thomas
Charles P. Felton National TB Center
2238 Fifth Ave., 1st Floor
New York, NY  10037
Purpose:  Treatment for LTBI has become a high priority in TB control.  Current clinical practice (CCP) for treatment of LTBI is self administered therapy.  A peer worker model offers an alternative modality.  This study compares cost-effectiveness of the peer monitor (PM) and CCP models in a randomized clinical trial in Harlem.

Methods:  A sample of 88 LTBI patients was randomly assigned to a PM or CCP group.  The PM group used peer workers, members of the same community who had completed TB treatment, to provide support and encourage adherence.  Information was collected on patients' utilization patterns and outcome measures.  Utilization cost data were derived from published sources.  The cost of the experimental intervention was assessed from budgets and imputed PM costs.  Cost-effectiveness was computed by dividing the difference between PM and CCP costs by the difference between PM and CCP outcomes (treatment completion and on-time treatment completion).

Results:  Completion of LTBI treatment was achieved by 88.9% of PM patients compared to 64.7% of CCP patients (p=0.013).  PM patients were also more likely to complete treatment on time (75.9% vs. 58.8% for controls; p=0.145).  Average annual costs were $9,340 per PM patient and $8,172 per CCP patient.  These differences were due to the additional program costs for PM patients, longer hospitalizations, and more clinic visits.  Compared to controls, the cost per additional completion of LTBI treatment for the PM group was $48, while the cost per additional on-time completion was $68.

Implications for Practice and Policy:  Our analysis suggests that the PM model is both more effective and more costly than CCP.  Comparing total costs and incremental costs for producing an additional completion, the PM strategy offers a cost-effective model for treatment of LTBI.